diff --git a/model_zoo/official/cv/ssd/src/dataset.py b/model_zoo/official/cv/ssd/src/dataset.py
index 19c66fc598..98097f474b 100644
--- a/model_zoo/official/cv/ssd/src/dataset.py
+++ b/model_zoo/official/cv/ssd/src/dataset.py
@@ -63,7 +63,7 @@ def random_sample_crop(image, boxes):
         if not drop_mask.any():
             continue
 
-        if overlap[drop_mask].min() < min_iou:
+        if overlap[drop_mask].min() < min_iou and overlap[drop_mask].max() > (min_iou + 0.2):
             continue
 
         image_t = image_t[rect[0]:rect[2], rect[1]:rect[3], :]
diff --git a/model_zoo/official/cv/ssd/src/init_params.py b/model_zoo/official/cv/ssd/src/init_params.py
index 6e1f8869b3..335030d2e9 100644
--- a/model_zoo/official/cv/ssd/src/init_params.py
+++ b/model_zoo/official/cv/ssd/src/init_params.py
@@ -14,16 +14,17 @@
 # ============================================================================
 """Parameters utils"""
 
-from mindspore import Tensor
 from mindspore.common.initializer import initializer, TruncatedNormal
+import numpy as np
 
 def init_net_param(network, initialize_mode='TruncatedNormal'):
     """Init the parameters in net."""
     params = network.trainable_params()
     for p in params:
-        if isinstance(p.data, Tensor) and 'beta' not in p.name and 'gamma' not in p.name and 'bias' not in p.name:
+        if 'beta' not in p.name and 'gamma' not in p.name and 'bias' not in p.name:
+            np.random.seed(seed=1)
             if initialize_mode == 'TruncatedNormal':
-                p.set_parameter_data(initializer(TruncatedNormal(0.03), p.data.shape, p.data.dtype))
+                p.set_parameter_data(initializer(TruncatedNormal(), p.data.shape, p.data.dtype))
             else:
                 p.set_parameter_data(initialize_mode, p.data.shape, p.data.dtype)